Page 1 of 19
Journal of Neurotrauma Formation of microparticles in the injured brain of patients with severe isolated traumatic brain injury. (doi: 10.1089/neu.2013.3168) This article has been peer-reviewed and accepted for publication, but has yet to undergo copyediting and proof correction. The final published version may differ from this proof.
1
Formation of microparticles in the injured brain of patients with severe isolated traumatic brain injury M.Nekludov1, F.Mobarrez2, D.Gryth1, B-M.Bellander3, H.Wallen2 1 – Karolinska Institutet, Department of Physiology and Pharmacology, section for Anesthesiology, Karolinska University Hospital 2 – Karolinska Institutet, Department of Clinical Sciences, Division for Cardiovascular Medicine, Danderyds Hospital 3 – Karolinska Institutet, Department of Clinical Neuroscience, section for Neurosurgery, Karolinska University Hospital.
Corresponding author: Dr.M.Nekludov, Dpt.Anesthesiology, Karolinska University Hospital Solna, 17176 Stockholm; [email protected]
Key words: Traumatic Brain Injury, Microparticles, Cerebrovenous, Coagulopathy, Neurosurgical ICU.
1
Page 2 of 19
Journal of Neurotrauma Formation of microparticles in the injured brain of patients with severe isolated traumatic brain injury. (doi: 10.1089/neu.2013.3168) This article has been peer-reviewed and accepted for publication, but has yet to undergo copyediting and proof correction. The final published version may differ from this proof.
2 Abstract The potential pathophysiological role of circulating microparticles (MPs) has been recognized in various conditions such as cardiovascular and thrombotic diseases. Traumatic brain injury has a complex pathophysiology which involves coagulopathy and inflammation. We investigated endothelial-, platelet- and leukocyte-derived microparticles (EMPs, PMPs and LMPs, respectively) in 16 patients with severe isolated TBI. Arterial and cerebrovenous samples were taken repeatedly, during 1-72 hours after the injury. Subpopulations of MPs, exposing Tissue Factor (TF) and P-selectin, were also studied. MP counts in cerebrovenous samples, irrespective of cellular origin, were higher in TBI cases, compared to healthy controls (peak levels of EMPs were around 7 times higher, PMPs 1.4 times higher, LMPs 2 times higher, respectively; p3 and non-head AIS ≤ 3 were considered to have severe isolated TBI.18 All patients underwent CT scan of the brain, and were included mean 6 hours (range 1-20 hours) after the trauma. Patients with known coagulation disorders, ongoing antithrombotic treatment with warfarin or platelet inhibiting drugs, or patients suffering from multiple trauma, were excluded.
Patient management The patients were treated and monitored in accordance with the local guidelines (based on Brain Trauma Foundation guidelines) at the NICU.19 An arterial line in the radial artery was established at arrival to the emergency room (ER). A jugular venous catheter was inserted in the jugular bulb shortly after the patient arrived to NICU, in order to monitor the jugulovenous oxygen saturation and to collect cerebrovenous blood samples. In acute trauma situations, this procedure may be delayed by other therapeutic and/or diagnostic measures of higher priority. In our material, the first cerebrovenous blood samples were therefore taken 9 hours (mean) after the injury (range 3-22 hours). The bleeding tendency was evaluated by a senior intensive care physician and a neurosurgeon, using a scoring system (no bleeding, moderate + and overt bleeding ++). 4
Page 5 of 19
Journal of Neurotrauma Formation of microparticles in the injured brain of patients with severe isolated traumatic brain injury. (doi: 10.1089/neu.2013.3168) This article has been peer-reviewed and accepted for publication, but has yet to undergo copyediting and proof correction. The final published version may differ from this proof.
5 Haemostatic treatments were given indicatively (Table 1). Patients received thrombosis prophylaxis with enoxaparin 20 mg subcutaneously daily, starting from day 2 after the injury. The investigators performing the present study were not involved in clinical decisions regarding the patients included in the study.
Measurements and blood sampling Blood samples were taken repeatedly and simultaneously from the arterial line and from the jugular bulb line at approximately 6, 12, 24, 48 and 72 hours after the trauma, respectively. The initial samples obtained at admission were taken from the arterial line only. All blood samples were drawn into a syringe containing citrate, pH 7.4 (1 part 0.129M trisodium citrate + 9 parts blood). The blood samples were centrifuged at 2500 x g in room temperature (RT) for 20 minutes immediately after sampling. The plasma obtained was divided into smaller tubes (0.4 mL), and kept frozen at -70ºC until the analysis. The MP analyses (see below) were performed in a two-step manner: first, “all MPs” irrespective of cellular origin were gated according to phosphatidylserine (PS) exposure and later phenotyped with monoclonal antibodies to detect cellular origin (endothelial-, plateletand leukocyte-derived MPs; EMPs, PMPs and LMPs). Thereafter we analyzed subpopulations of MPs, i.e. those exposing tissue factor (TF), and in PMPs also the expression of the adhesion molecule P-selectin (CD62P).
Analysis of circulating microparticles The previously frozen platelet poor plasma was thawed and centrifuged at 2000 x g for 20 minutes at RT. The supernatant was re-centrifuged, at 13000 x g for 2 minutes at RT. Twenty µL of the supernatant was then incubated for 20 minutes in dark with phalloidin-Alexa-660 (Invitrogen, Paisley, UK) lactadherin-FITC (Haematologic Technologies, Vermont, USA), and monoclonal antibodies to detect cell origin: anti-CD42a-PE (BD, NJ, USA) to detect platelet origin, anti-CD144-APC (AH diagnostics, Stockholm, SWE) to detect endothelial origin, and anti-CD45 (Beckman Coulter, Brea, Ca, USA) to detect leukocyte origin. Pselectin was detected with anti-CD62P-PE (BD, NJ, USA) and TF with anti-CD142-PE (clone HTF-1 IgG1κ, BD, NJ, USA).20 MPs were measured by flow cytometry on a Beckman Gallios instrument (Brea, CA, USA) and identified by size (< 1.0 µm in size) and binding of fluorescent labeled lactadherin and the labeled antibodies as described above. The MP-gate was determined using Megamix beads (BioCytex, Marseille, France), which is a mix of beads with diameters of 0.5 µm, 0.9 µm and 5
Page 6 of 19
Journal of Neurotrauma Formation of microparticles in the injured brain of patients with severe isolated traumatic brain injury. (doi: 10.1089/neu.2013.3168) This article has been peer-reviewed and accepted for publication, but has yet to undergo copyediting and proof correction. The final published version may differ from this proof.
6 3.0 µm, respectively. MPs were defined as particles less than 1.0 µm in size, negative to phalloidin (in order to exclude cell membrane fragments; see 21) and positive to lactadherin. Notably, lactadherin instead of annexinV, was used to identify phosphatidylserine exposure on MPs, as lactadherin is a better probe than annexin V in detecting PS expressing particles.22 Conjugate isotype-matched immunoglobulin (IgG1-FITC, IgG1-PE and IgG1-APC) with no reactivity against human antigens was used as a negative control to define the background noise of the cytometry analysis. The absolute number of MPs was calculated by means of the following formula: (MP counted x standard beads ⁄ L) ⁄ standard beads counted, (FlowCount, Beckman Coulter).
Statistics Statistical analyses were performed using Statistica v.10 (StatSoft Inc, USA) and SPSS Statistics software (IBM Software). Data are presented as mean ±standard deviations (SD). We analysed the data for the presence of time trends between the sampling points (falling time trend), and for the difference between the arterial and cerebrovenous values (transcranial gradient). This was performed using the Mixed model analysis. This model accommodates missing data, assuming that data are missing at random. Post hoc test were performed within the mixed model analysis, using the Sidak correction for multiple comparisons. Importantly, the results obtained with the mixed model procedure were in agreement with those obtained with the traditional t-test procedure for repeated measurements of nonparametric data (Wilcoxon test). Data were considered to be normally distributed if -1 < skewness < 1. This was true for all MP data included in the Mixed models analysis, except at the 48 h measurement for CD42a/142 and CD42a/62P where skewness was marginally outside these limits (i.e. 1.08 and 1.05, respectively). Graph Pad program 6.0 (GraphPad Software Inc, USA) was used for presentation of the results.
Results Sixteen adult patients (mean age 45 years; four women and 12 men) were studied. Reference values for MP counts in venous plasma were obtained from a control group of 15 healthy volunteers (7 men, 8 women, mean age 44 years). Patient characteristics are shown in Table 1.
6
Page 7 of 19
Journal of Neurotrauma Formation of microparticles in the injured brain of patients with severe isolated traumatic brain injury. (doi: 10.1089/neu.2013.3168) This article has been peer-reviewed and accepted for publication, but has yet to undergo copyediting and proof correction. The final published version may differ from this proof.
7 MPs of various cellular origins regardless of CD142 or CD62P exposure, (Figure 1) Plasma levels of EMPs (CD144+ particles), PMPs (CD42a+ particles) and LMPs (CD45+ particles) in the first samples obtained were higher in TBI patients than in healthy controls (4.1±1.8 x109 vs 0.5±0.3 x109/L, 15.3±9.8 x109 vs 9.9±2.6 x109/L and 4.9±1.4 x109 vs 2.1±0.6 x109/L, respectively, p
Journal of Neurotrauma Formation of microparticles in the injured brain of patients with severe isolated traumatic brain injury. (doi: 10.1089/neu.2013.3168) This article has been peer-reviewed and accepted for publication, but has yet to undergo copyediting and proof correction. The final published version may differ from this proof.
1
Formation of microparticles in the injured brain of patients with severe isolated traumatic brain injury M.Nekludov1, F.Mobarrez2, D.Gryth1, B-M.Bellander3, H.Wallen2 1 – Karolinska Institutet, Department of Physiology and Pharmacology, section for Anesthesiology, Karolinska University Hospital 2 – Karolinska Institutet, Department of Clinical Sciences, Division for Cardiovascular Medicine, Danderyds Hospital 3 – Karolinska Institutet, Department of Clinical Neuroscience, section for Neurosurgery, Karolinska University Hospital.
Corresponding author: Dr.M.Nekludov, Dpt.Anesthesiology, Karolinska University Hospital Solna, 17176 Stockholm; [email protected]
Key words: Traumatic Brain Injury, Microparticles, Cerebrovenous, Coagulopathy, Neurosurgical ICU.
1
Page 2 of 19
Journal of Neurotrauma Formation of microparticles in the injured brain of patients with severe isolated traumatic brain injury. (doi: 10.1089/neu.2013.3168) This article has been peer-reviewed and accepted for publication, but has yet to undergo copyediting and proof correction. The final published version may differ from this proof.
2 Abstract The potential pathophysiological role of circulating microparticles (MPs) has been recognized in various conditions such as cardiovascular and thrombotic diseases. Traumatic brain injury has a complex pathophysiology which involves coagulopathy and inflammation. We investigated endothelial-, platelet- and leukocyte-derived microparticles (EMPs, PMPs and LMPs, respectively) in 16 patients with severe isolated TBI. Arterial and cerebrovenous samples were taken repeatedly, during 1-72 hours after the injury. Subpopulations of MPs, exposing Tissue Factor (TF) and P-selectin, were also studied. MP counts in cerebrovenous samples, irrespective of cellular origin, were higher in TBI cases, compared to healthy controls (peak levels of EMPs were around 7 times higher, PMPs 1.4 times higher, LMPs 2 times higher, respectively; p3 and non-head AIS ≤ 3 were considered to have severe isolated TBI.18 All patients underwent CT scan of the brain, and were included mean 6 hours (range 1-20 hours) after the trauma. Patients with known coagulation disorders, ongoing antithrombotic treatment with warfarin or platelet inhibiting drugs, or patients suffering from multiple trauma, were excluded.
Patient management The patients were treated and monitored in accordance with the local guidelines (based on Brain Trauma Foundation guidelines) at the NICU.19 An arterial line in the radial artery was established at arrival to the emergency room (ER). A jugular venous catheter was inserted in the jugular bulb shortly after the patient arrived to NICU, in order to monitor the jugulovenous oxygen saturation and to collect cerebrovenous blood samples. In acute trauma situations, this procedure may be delayed by other therapeutic and/or diagnostic measures of higher priority. In our material, the first cerebrovenous blood samples were therefore taken 9 hours (mean) after the injury (range 3-22 hours). The bleeding tendency was evaluated by a senior intensive care physician and a neurosurgeon, using a scoring system (no bleeding, moderate + and overt bleeding ++). 4
Page 5 of 19
Journal of Neurotrauma Formation of microparticles in the injured brain of patients with severe isolated traumatic brain injury. (doi: 10.1089/neu.2013.3168) This article has been peer-reviewed and accepted for publication, but has yet to undergo copyediting and proof correction. The final published version may differ from this proof.
5 Haemostatic treatments were given indicatively (Table 1). Patients received thrombosis prophylaxis with enoxaparin 20 mg subcutaneously daily, starting from day 2 after the injury. The investigators performing the present study were not involved in clinical decisions regarding the patients included in the study.
Measurements and blood sampling Blood samples were taken repeatedly and simultaneously from the arterial line and from the jugular bulb line at approximately 6, 12, 24, 48 and 72 hours after the trauma, respectively. The initial samples obtained at admission were taken from the arterial line only. All blood samples were drawn into a syringe containing citrate, pH 7.4 (1 part 0.129M trisodium citrate + 9 parts blood). The blood samples were centrifuged at 2500 x g in room temperature (RT) for 20 minutes immediately after sampling. The plasma obtained was divided into smaller tubes (0.4 mL), and kept frozen at -70ºC until the analysis. The MP analyses (see below) were performed in a two-step manner: first, “all MPs” irrespective of cellular origin were gated according to phosphatidylserine (PS) exposure and later phenotyped with monoclonal antibodies to detect cellular origin (endothelial-, plateletand leukocyte-derived MPs; EMPs, PMPs and LMPs). Thereafter we analyzed subpopulations of MPs, i.e. those exposing tissue factor (TF), and in PMPs also the expression of the adhesion molecule P-selectin (CD62P).
Analysis of circulating microparticles The previously frozen platelet poor plasma was thawed and centrifuged at 2000 x g for 20 minutes at RT. The supernatant was re-centrifuged, at 13000 x g for 2 minutes at RT. Twenty µL of the supernatant was then incubated for 20 minutes in dark with phalloidin-Alexa-660 (Invitrogen, Paisley, UK) lactadherin-FITC (Haematologic Technologies, Vermont, USA), and monoclonal antibodies to detect cell origin: anti-CD42a-PE (BD, NJ, USA) to detect platelet origin, anti-CD144-APC (AH diagnostics, Stockholm, SWE) to detect endothelial origin, and anti-CD45 (Beckman Coulter, Brea, Ca, USA) to detect leukocyte origin. Pselectin was detected with anti-CD62P-PE (BD, NJ, USA) and TF with anti-CD142-PE (clone HTF-1 IgG1κ, BD, NJ, USA).20 MPs were measured by flow cytometry on a Beckman Gallios instrument (Brea, CA, USA) and identified by size (< 1.0 µm in size) and binding of fluorescent labeled lactadherin and the labeled antibodies as described above. The MP-gate was determined using Megamix beads (BioCytex, Marseille, France), which is a mix of beads with diameters of 0.5 µm, 0.9 µm and 5
Page 6 of 19
Journal of Neurotrauma Formation of microparticles in the injured brain of patients with severe isolated traumatic brain injury. (doi: 10.1089/neu.2013.3168) This article has been peer-reviewed and accepted for publication, but has yet to undergo copyediting and proof correction. The final published version may differ from this proof.
6 3.0 µm, respectively. MPs were defined as particles less than 1.0 µm in size, negative to phalloidin (in order to exclude cell membrane fragments; see 21) and positive to lactadherin. Notably, lactadherin instead of annexinV, was used to identify phosphatidylserine exposure on MPs, as lactadherin is a better probe than annexin V in detecting PS expressing particles.22 Conjugate isotype-matched immunoglobulin (IgG1-FITC, IgG1-PE and IgG1-APC) with no reactivity against human antigens was used as a negative control to define the background noise of the cytometry analysis. The absolute number of MPs was calculated by means of the following formula: (MP counted x standard beads ⁄ L) ⁄ standard beads counted, (FlowCount, Beckman Coulter).
Statistics Statistical analyses were performed using Statistica v.10 (StatSoft Inc, USA) and SPSS Statistics software (IBM Software). Data are presented as mean ±standard deviations (SD). We analysed the data for the presence of time trends between the sampling points (falling time trend), and for the difference between the arterial and cerebrovenous values (transcranial gradient). This was performed using the Mixed model analysis. This model accommodates missing data, assuming that data are missing at random. Post hoc test were performed within the mixed model analysis, using the Sidak correction for multiple comparisons. Importantly, the results obtained with the mixed model procedure were in agreement with those obtained with the traditional t-test procedure for repeated measurements of nonparametric data (Wilcoxon test). Data were considered to be normally distributed if -1 < skewness < 1. This was true for all MP data included in the Mixed models analysis, except at the 48 h measurement for CD42a/142 and CD42a/62P where skewness was marginally outside these limits (i.e. 1.08 and 1.05, respectively). Graph Pad program 6.0 (GraphPad Software Inc, USA) was used for presentation of the results.
Results Sixteen adult patients (mean age 45 years; four women and 12 men) were studied. Reference values for MP counts in venous plasma were obtained from a control group of 15 healthy volunteers (7 men, 8 women, mean age 44 years). Patient characteristics are shown in Table 1.
6
Page 7 of 19
Journal of Neurotrauma Formation of microparticles in the injured brain of patients with severe isolated traumatic brain injury. (doi: 10.1089/neu.2013.3168) This article has been peer-reviewed and accepted for publication, but has yet to undergo copyediting and proof correction. The final published version may differ from this proof.
7 MPs of various cellular origins regardless of CD142 or CD62P exposure, (Figure 1) Plasma levels of EMPs (CD144+ particles), PMPs (CD42a+ particles) and LMPs (CD45+ particles) in the first samples obtained were higher in TBI patients than in healthy controls (4.1±1.8 x109 vs 0.5±0.3 x109/L, 15.3±9.8 x109 vs 9.9±2.6 x109/L and 4.9±1.4 x109 vs 2.1±0.6 x109/L, respectively, p
